Protegrity’s AI Friction Tax Research Featured in VMblog

Organizations can build capable AI applications and still encounter significant barriers before those systems reach production. Security and compliance reviews can delay deployments, while concerns about sensitive data can lead teams to reduce what their AI systems are allowed to do.

In its October 30, 2026 roundup, Scary Cybersecurity Stats: Alarming Findings from 2026 Research, VMblog highlights findings from Protegrity's AI Friction Tax research. Conducted by Enterprise Management Associates (EMA) and commissioned by Protegrity, the survey of 152 IT and security leaders examines how security, governance, and compliance requirements are affecting enterprise AI deployment.

Security Reviews Are Delaying Production AI

According to the research cited by VMblog, 82.9% of surveyed organizations reported experiencing AI production delays due to security or compliance reviews.

Among respondents who experienced delays, 66.7% reported setbacks lasting one month or longer.

These findings illustrate a challenge for enterprise AI teams. Developing an application is only one part of reaching production. Organizations must also address how sensitive information is accessed, protected, and governed before AI applications can operate within approved business environments.

Security Concerns Are Limiting AI Capabilities

Delays are not the only consequence highlighted in the research. The survey also found that 81.6% of respondents reported deploying AI with diminished capabilities because of security concerns.

The VMblog roundup highlights another finding: 62.5% of respondents identified data leakage through autonomous agent actions as their top security concern.

Together, these findings help explain why organizations may restrict access to sensitive information or limit agent functionality when adequate protection and governance controls are not in place.

Building Data Protection Into AI Workflows

The research also points to the importance of integrating data protection directly into AI operations.

VMblog reports that 91.5% of surveyed leaders considered a solution that embeds data protection into AI workflows extremely or very valuable. Additionally, 47.4% identified reducing manual audit and compliance costs as their top desired business outcome from improved AI data security and governance.

These results reinforce the need to address data protection and governance throughout the AI development and deployment process, rather than treating them as separate requirements at the final review stage.

For organizations evaluating this approach, Protegrity's data protection capabilities help protect sensitive information while supporting authorized use across data and AI workflows.

This article summarizes the Protegrity research findings referenced in VMblog's "Scary Cybersecurity Stats: Alarming Findings from 2026 Research." Read the original publication for the full collection of cybersecurity research and industry findings.